Large Charset Problem

Hi I've the following problem:
We have an Oracle 10 database running on Red Hat Linux:
Now we also have a tomcat 5 on this Linux-Server, but our web-application cannot display "umlaute" like
Ã or Ã correctly.It cannot depend on the database, because we use it also from
another Tomcat installed on Windows and there the symbols will be displayed correctly.
It also cannot depend on the configuration of the Tomcat 5 because the error already exists in the standard-output (and not appears just in the HTML encoding).
We assume that it is the oracle driver, but we use the same both on the windows- and on the linux-tomcat.
So the only difference are left is the difference between linux and windows.
Has anybody a suggestion what we could do?
Thank you very much
Christian

By "client" here I mean the middle tier (the machine where the Oracle client resides).
The format of the NLS_LANG environment variable is
<<language>>_<<territory>>.<<character set>>So
AMERICAN_AMERICA.WE8MSWIN1252
GERMAN_GERMANY.WE8ISO8859P15
It looks like you may have a period rather than an unerscore between language & territory.
Justin
Distributed Database Consulting, Inc.
http://www.ddbcinc.com/askDDBC

Similar Messages

[Gnome - GDM] Locale/Charset Problem

Hi everybody,
I´m using Gnome with GDM as Display Manager.
I got a problem with my keyboard settings, but don´t know why
The first charset/locale option is set in the /boot/grub/menu.lst
kernel /vmlinuz26 lang=de locale=de_DE.UTF8 root=/dev.....
I use a LUKS encrypted /home partition and need a german locale/charset for the passphrase/during bootup.
The second charset option is in my /etc/rc.conf:
LOCALE="de_DE.UTF8"
KEYMAP="de"
With those settings I got an almost "normal" keyboard behaviour.
The Umlaut-Character for example arent´t working during bootup or in a tty. In a later started shell inside of Gnome everything works fine.
The second charset problem is during password input inside GDM. The layout isn´t working right here either.
I already tried the ./dmrc settings of Language and Keys, but nothing works.
Anyone got a clue how to fix that?
Greets
Flo

Maybe you can change the subject of this thread, as the issue is not only related to gdm, IMHO.
I don't get it either. Settings are LOCALE="sv_SE" and KEYMAP="sv-latin1" in rc.conf and locale -a gives me
[root@localhost ~]# locale -a
C
POSIX
sv_SE
sv_SE.iso88591
swedish
But something is still wrong, because calling up some manpages gives me
Cannot open the message catalog "man" for locale "sv_SE"
(NLSPATH="<none>")
I ran locale-gen, but I'm pretty lost in this locale business too...

Informix 2 Oracle... charset problem?

Hi,
i'm tring to migrate data from informix 9 to oracle 9.2 but while migrating 2 tables i 've got this errors:
ORA-01401: inserted value too large for column
and
ORA-01461: can bind a LONG value only for insert into a LONG column
The same procedure works ok if i migrate data to another oracle db server... the differences between this 2 oracle are:
WORKS:
NLS_RDBMS_VERSION= 10.2.0.1.0
NLS_CHARACTERSET= WE8MSWIN1252
NOT WORKS
NLS_CHARACTERSET=UTF8
NLS_RDBMS_VERSION=9.2.0.5.0
Any suggestions? is only a problem of characterset??

I reckon thats exactly what it is, a problem of charset and the storage requirements of the data therein. do you know which table in the repository is having this problem inserting data?
Message was edited by:
Barry McGillin

Copying large files problem in windows 7 x64 professional

Hi!
I'm new to this forum and I recently installed windows 7 professional without too much trouble in my pc. The only thigs that doesn't work are the quicklaunch buttons (which I'm expecting HP to release it until october 22). The system is functioning ok, but I discovered a very serious problem when tranferring large size files (copy or move function) to an external source (in my case a external HD: model Western Digital My book 1 TB connected via USB).
When I try to move a file that is larger than 1 GB the system during the process gets slower and slower rendering my system useless unless I cancel the process and restart. This only happens when moving files from my notebook to my HD.
I've been doing my homework and I checked microsoft technet forums, here's what I found:
1.It seems a lot of people have this problem, in particullar the ones with an asus motherboard (which I know this isn't the case). But in my case since I don't know which motherboard my notebook has, I can't come up with a solution to this problem.
2. It seems to be a particullar problem with the 64 version of windows.
3. The users claim that the problem was solved when they upgraded the chipset, which I somehow tried but with the current vista drivers, but the problem wasn't solved.
4. Some users claim that by using other software like teracopy or grant full permission control seems like a workaround to this, but no in my case. It doesn't work.
5. With sisoft sandra I managed to get my motherboard from this notebook but with no luck finding information or upgrades to a Quanta motherboard.
Here are my system specs (link to the HP site describing it):
Model: HP pavilion 6950la
http://h10025.www1.hp.com/ewfrf/wc/document?docname=c01470050&lc=en&dlc=en&cc=us&lang=en&softwareitem=ob-55781-1&os=2100&product=3753762
Windows 7 professional x64
I'm worried about this and it seems weird to me that no one who has tried windows 7 professional x64 has mentioned this in the forum (or even mentioning it), because I think this is a really serious issue and I would like to make sure it gets known and hopefully solved during these days. I really don't want ot get back to vista just for this. And I hope my post will be useful to the HP people so they can get around this problem.
Thankd in advance, will wait for your answers.

Please repost your inquiry in the Microsoft Windows 7 Performance Forum.
Thank you!
Carey Frisch
Microsoft MVP
Windows Expert - Consumer

Airport Extreme, Airdisk, and Vista - large file problem with a twist

Hi all,
I'm having a problem moving large-ish files to an external drive attached to my Airport Extreme SOMETIMES. Let me explain.
My system - Macbook Pro, 4gb ram, 10gb free HD space on macbook, running latest updates for mac and vista, external hard drive on AE is an internal WD in an enclosure w/ 25gb free (its formatted for pc, but I've used it directly connected to multiple computers, mac and pc, without fail), AE is using firmware 7.3.2, and im only allowing 802.11n. The problem occurs on the Vista side - havent checked the mac side yet.
The Good - I have bit torrents set up, using utorrent, to automatically copy files over to my airdisk once they've completed. This works flawlessly. If i connect the hard drive directly to my laptop (macbook running bootcamp w/ vista, all updates applied), large files copy over without a hitch as well.
The Bad - For the past couple weeks (could be longer, but I've just noticed it recently being a problem - is that a firmware problem clue?) If i download the files to my Vista desktop and copy them over manually, the copy just sits for a while and eventually fails with a "try again" error. If i try to copy any file over 300mb, the same error occurs.
What I've tried - well, not a lot. The first thing i did was to make sure my hard drive was error free, and worked when physically connected - it is and it did. I've read a few posts about formatting the drive for mac, but this really isnt a good option for me since all of my music is on this drive, and itunes pulls from it (which also works without a problem). I do however get the hang in itunes if i try to import large files (movies). I've also read about trying to go back to an earlier AE firmware, but the posts were outdated. I can try the mac side and see if large files move over, but again, i prefer to do this in windows vista.
this is my first post, so im sure im leaving out vital info. If anyone wants to take a stab at helping, I'd love to discuss. Thanks in advance.

Hello,
Just noticed the other day that I am having the same problem. I have two Vista machines attached to TC w/ a Western Digital 500 gig USB'd. I can write to the TC (any file size) with no problem, however I cannot write larger files to my attached WD drive. I can write smaller folders of music and such, but I cannot back up larger video files. Yet, if I directly attach the drive to my laptop I can copy over the files, no problem. I could not find any setting in the Airport Utility with regards to file size limits or anything of the like. Any help on this would be much appreciated.

Mail attachment charset problem

Hello,
I have made a program which is able to send Icalendar files as an attachment. I get the data as an InputStream.
My problem is that the Icalendar file doesn�t show the letters '�', '�' and '�'. I have tried to use iso-8859-1 in the MimeBodyPart headerline and in the ByteArrayDataSource, but it doesn�t work?!
Where can I specify which charset I want to use?
MimeBodyPart mbp3 = new MimeBodyPart();
mbp3.setFileName( m.getAttachmentFileName() );
mbp3.setHeader("Content-Class", "urn:content-classes:calendarmessage");
mbp3.setHeader("Content-ID","calendar_message");
mbp3.addHeaderLine("charset=iso-8859-1");
java.io.InputStream inputStream = null;
try {
inputStream = m.getAttachmentFile().getBinaryStream();
mbp3.setDataHandler( new DataHandler( new javax.mail.util.ByteArrayDataSource( inputStream, "text/calendar;charset=iso-8859-1;method=REQUEST" ) ) );
catch ( Exception e ){}
mpRoot.addBodyPart(mbp3);

Yes you are right... Thank you.
I removed the line:
mbp3.addHeaderLine("charset=iso-8859-1"); - and now the letters are shown correct when opening the ICalendar file using a text editor.
But when openning the file using Outlook the letters '�', '�', '�' are removed?! I know that isn�t a problem in my mail code but certainly in the iCal file?!

JSP, Javabean charset problem

I have some JSP pages where I try to dynamically present some drop-down
menus for the users to select values. I use a simple bean to manage it.
The problem is that those values are in non-iso8859-1 charset and I only
get ?????? rendered in the select box. I define an array (inline in the
JSP page as code scriptlet), write all possible (String) options for the
drop-down menu there and in the bean I do some calculations and render
the drop-down menu.
String label[]={"something in iso-8859-7 encoding in here","something in
iso-8859-7 encoding in here","something in iso-8859-7 encoding in here"};
and in the bean I have a for-loop to access this.
The page directive is set to iso-8859-7.
I think there is some kind of transparent translation, that has to do
with Java language, and after the rendering I only get ???? instead of
the correct iso-8859-7 value in the browser.
Any help appreciated.
(Tomcat, Apache web server, JDK 1.3,)
PS: This JSP page is used to submit some data in an Oracle database
(according to the selection of the user in the drop-down box), so I also
use JDBC 1.3, but I don't think that's relevant at all with my problem...
null

I have some JSP pages where I try to dynamically present some drop-down
menus for the users to select values. I use a simple bean to manage it.
The problem is that those values are in non-iso8859-1 charset and I only
get ?????? rendered in the select box. I define an array (inline in the
JSP page as code scriptlet), write all possible (String) options for the
drop-down menu there and in the bean I do some calculations and render
the drop-down menu.
String label[]={"something in iso-8859-7 encoding in here","something in
iso-8859-7 encoding in here","something in iso-8859-7 encoding in here"};
and in the bean I have a for-loop to access this.
The page directive is set to iso-8859-7.
I think there is some kind of transparent translation, that has to do
with Java language, and after the rendering I only get ???? instead of
the correct iso-8859-7 value in the browser.
Any help appreciated.
(Tomcat, Apache web server, JDK 1.3,)
PS: This JSP page is used to submit some data in an Oracle database
(according to the selection of the user in the drop-down box), so I also
use JDBC 1.3, but I don't think that's relevant at all with my problem...
null

JDev 1013 EA1 : embeded OC4J default-charset problem

Hi
I'm a web-application programmer interesting in JDeveloper.
I had tried development using JDeveloper 10.1.2
When I used 10.1.2, I add [default-charset="EUC-KR"] to [JDEV_HOME/systemXX/config/global-application.xml]
- or [deployment-application/XXX/XXX/orion-web.xml] -
Because I don't use alphabet.
Anyway, it have effect.
But I encountered problem, after use JDeveloper 10.1.3 EA1.
I set default-charset in embeded-OC4J.
But it had no effect.
If that is a bug, please tell me how to use a installed OC4J in JDeveloper.
Please help me.
Thanks for read this message
Message was edited by:
user453010

This looks like a bug in OC4J
Can you try this and let us know if it works?
Set the charset in the application level orion-web.xml and deploy it to the Stand alone OC4J and see if it works.
1. Add a new OC4J Deployment Descriptor to your project
2. Choose orion-web.xml
3. Open the file in editor
4. Set the default-charset setting
<orion-web-app xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:noNamespaceSchemaLocation="http://xmlns.oracle.com/oracleas/schema/orion-web-10_0.xsd"
schema-major-version="10" schema-minor-version="0"
servlet-webdir="/servlet/" default-charset="EUC-KR"></orion-web-app>
(Make sure all the other settings are set right. Similar to what you see in [deployment-application/XXX/XXX/orion-web.xml] )
5. Create a WAR deployment profile
6. Deploy the application to Standalone OC4J connection

Prime Infrastructure 2.0 email charset problem

Dear all,
I have question regarding Guest portal managed through PI. Everything is set, but only one thing doesn't work as I expected.
When I make an account and print it, everything looks fine. When I sent it via email, then enconding of mail is wrong.
Problem is, that PI gives Charset : ANSI_X3.4-1968 (can see in email header).
I would like to have ther (in email) : CP1252 (or UTF) - basically something what can interpret national characters.
I have found this about similar problem, but absolutely have no idea where set it up.
http://thwack.solarwinds.com/thread/58101
http://helpdesk.ibs-aachen.de/2006/05/13/javamail-error-in-subject-ansi_x34-1968qtestmail_-_best3ftigung_ihrer_email-adresse/
In old WCS, there was no problem with this.
Does anyone has solution for this?
Thank you
Pavel

I see now, this is a poor carry over from NCS. The option that says include "controller" in email does provide the switch hostname. Hopefully by version 3.0 they will have all the wireless naming fixed when it applies to converged.
Thanks Rob!

Charset problems

Have also problems with charset.
Installed German translation and IdM doesn't save German special types in my MYSQL DB but shows me cryptic types.
When deleting a user in the english Admin site there's no problem but IdM shows me an errors message when deleting a user in the german surface. Also the user is deleted IdM isn't able to assign the special german fonts...

Hi,
i migrated a production environment with 40k userIds and 160k accounts from mysql 4.0.x to 4.1.x two weeks ago (unsure about the exact subversions atm).
We did have some issues (unable to delete IDM orgs) about collations as there were mayor characterset and collation changes between those mysql versions. This could be fixed with a few alter tables tho.
Oh and i did forget to mention before that 80% of our 380 admins are located in germany just as the servers are. So there are plenty of ��.
As it is a production environment dozens of users get deleted on a dayly basis and i didn't encounter your problem. What jdbc driver (version) are you using?
Regards,
Patrick

JMS Sender SAP JMS Provide large message problem

Hi,
we have configured a JMS sender channel to pick up messages from a queue hosted by our SAP JMS provider. Unfortunately a message of about 6 MB size isn't picked up. Smaller messages are picked up.
Has anybody experienced a problem like this before?
Kind regards,
Heiko

Hey,
I guess you are using XML messages:
We had some problems with large XML messages (e.g more the 10MB),
This is usally cause by memory problems.
There are some work-around, like increasing the memory usage of the application server.
The first thing you do is try to understand where the message stacked (in the ABAP/J2EE).
If this is a memory problem, changing the memory configuration can improve this,
but be aware that there are hardware limitation (32bit application server can use with one process only 2GB), therefore messages over 100MB probably will not pass trough XI.
If you have huge files (e.g more than 100MB)
You must develop a program that would split the large message into several small messages. the program can not be written in the XI, and should be written before the adapter. (you might install and use in this case the conversion agent)
If you are using a message with csv format (not XML),
than it is possible to configure the adapter to split every X lines
(no program need to be written)

Charset problem!!

Hello everybody, there is something i need help with..
I have xml files, created by me in java, i create xml files in different charsets. The problem is when i try to read a UTF-8 File, with :
StreamSource sts=new StreamSource(new BufferedReader(new InputStreamReader(new FileInputStream(filename),"UTF-8")));when i read a ISO-8859-1 file, it works fine, but with a UTF-8 is seems to be a problem, cause i get a error parsing the xml, but if i use :
StreamSource sts=new StreamSource(filename);it works for UTF-8, but not for ISO, (i think UTF-8 is java's default)..
so, any ideas??

If you pass the XML parser an InputStream or
something else where it can read the bytes, it will
get things right. (Assuming the XML declares its
encoding correctly.) If you pass the XML parser a
Reader where you apply the wrong encoding, the parser
will not be able to get things right. Choose one of
those two.That's why i used:
StreamSource sts=new StreamSource(new BufferedReader(new InputStreamReader(new FileInputStream(filename),"UTF-8")));I read that InputStreamReader is the best way to read characters well using a given encoding.

WebService response charset problems

Some time ago I posted a problem regarding unicode characters in WebService
responses from WL7.0. I received a patch that helped partially - after the
patch was applied, WL no longer threw any exception when an utf character
was included in the response. So far so good.
However, a problem arises when I call the WebService from a .NET client;
.NET doesn't understand that the response is utf-8 encoded, so when the
response is deserialized on the client side the encoded characters (such as
å, ä, ö) come out as question marks. It seems that the Content-Type header
doesn't specify the correct charset (I would expect something like
'Content-Type:text/xml; charset=utf-8', but the charset=... part seems to be
missing)
By fiddling about a bit with the .NET generated proxy class I managed to
force .NET to think that the Content-Type mime header does in fact contain
the correct value (quite messy - I can supply the code if anyone should be
interested). However, this should not be necessary - the solution I came up
with is awkward and the only thing needed is that the correct Content-Type
header be included in the WebService response. Is there a way to specify a
default value for this?
I tried creating a handler to intercept the response and set this specific
mime header, but no luck - the value I set seems to be ignored (i tried
ctx.getMessage().getMimeHeaders().setHeader("Content-Type", "text/xml;
charset=utf-8");, as well as ...addHeader()). Besides, even if this did work
it would seem unnecessarity complicated to create a handler and set it to
handle all the methods in my WebService (there are quite a few).
Any ideas?
/Mattias Arthursson
Compost Marketing

This problem should be fixed in SP1. If the system property
user.lang is not english, then SP1 will use utf-8 as charset
(I think this will be your case).
In SP1 you can also set a system property to change charset :
weblogic.webservice.i18n.charset="my-char-set"
regards,
-manoj
"Mattias Arthursson" <[email protected]> wrote in message
news:[email protected]...
Some time ago I posted a problem regarding unicode characters inWebService
responses from WL7.0. I received a patch that helped partially - after the
patch was applied, WL no longer threw any exception when an utf character
was included in the response. So far so good.
However, a problem arises when I call the WebService from a .NET client;
.NET doesn't understand that the response is utf-8 encoded, so when the
response is deserialized on the client side the encoded characters (suchas
å, ä, ö) come out as question marks. It seems that the Content-Type header
doesn't specify the correct charset (I would expect something like
'Content-Type:text/xml; charset=utf-8', but the charset=... part seems tobe
missing)
By fiddling about a bit with the .NET generated proxy class I managed to
force .NET to think that the Content-Type mime header does in fact contain
the correct value (quite messy - I can supply the code if anyone should be
interested). However, this should not be necessary - the solution I cameup
with is awkward and the only thing needed is that the correct Content-Type
header be included in the WebService response. Is there a way to specify a
default value for this?
I tried creating a handler to intercept the response and set this specific
mime header, but no luck - the value I set seems to be ignored (i tried
ctx.getMessage().getMimeHeaders().setHeader("Content-Type", "text/xml;
charset=utf-8");, as well as ...addHeader()). Besides, even if this didwork
it would seem unnecessarity complicated to create a handler and set it to
handle all the methods in my WebService (there are quite a few).
Any ideas?
/Mattias Arthursson
Compost Marketing

Charset problems using SQLLDR

Hello!
My task is to import data from a Microsoft Access Database into an Oracle Database. I got a Script which creates Flatfiles and Controlfiles with the data from Access. My problem is the characterset. There are some signs (e.g. the typical german quotationmark at the BOTTOM oder the "longer dash" you get from MS Word by using "blank+dash+blank"), which obviously cannot be convertet. So there are "turn-round questionsmarks". So I tried many different char-sets. I tried different nls_charset settings at oracle side and many charset as the parameter of the control files. But none combination let me get the right result. Does anybody know how I can get rid of this consersation desaster?
Best regards, Sascha Meyer

Hello!
Of cource I can give you the information:
Code points:
201C, 201D, 201E, 2013,2014
nls_database_parameters:
NLS_TIME_FORMAT HH24:MI:SSXFF
NLS_TIMESTAMP_FORMAT DD.MM.RR HH24:MI:SSXFF
NLS_TIME_TZ_FORMAT HH24:MI:SSXFF TZR
NLS_TIMESTAMP_TZ_FORMAT DD.MM.RR HH24:MI:SSXFF TZR
NLS_DUAL_CURRENCY ?
NLS_COMP BINARY
NLS_LENGTH_SEMANTICS BYTE
NLS_NCHAR_CONV_EXCP FALSE
NLS_NCHAR_CHARACTERSET AL16UTF16
NLS_RDBMS_VERSION 10.2.0.2.0
nls_session_parameter:
PARAMETER VALUE
NLS_LANGUAGE GERMAN
NLS_TERRITORY GERMANY
NLS_CURRENCY
NLS_ISO_CURRENCY GERMANY
NLS_NUMERIC_CHARACTERS ,.
NLS_CALENDAR GREGORIAN
NLS_DATE_FORMAT DD.MM.RR
NLS_DATE_LANGUAGE GERMAN
NLS_SORT GERMAN
NLS_TIME_FORMAT HH24:MI:SSXFF
NLS_TIMESTAMP_FORMAT DD.MM.RR HH24:MI:SSXFF
NLS_TIME_TZ_FORMAT HH24:MI:SSXFF TZR
NLS_TIMESTAMP_TZ_FORMAT DD.MM.RR HH24:MI:SSXFF TZR
NLS_DUAL_CURRENCY
NLS_COMP BINARY
NLS_LENGTH_SEMANTICS BYTE
NLS_NCHAR_CONV_EXCP FALSE
Default value is WE8ISO8859P15
I tried this setting:
Database: WE8ISO8859P15
Charset used in controlfile: WE8MSWIN1252, UTF8, WE8ISO8859P15, we8iso8859p1, AL32UTF8
Database: UTF8
Charset used in controlfile:WE8ISO8859P15, UTF8, we8iso8859p1, AL32UTF8
Database:WE8MSWIN1252
Charset used in controlfile:WE8ISO8859P15, UTF8, WE8MSWIN1252, we8iso8859p1, AL32UTF8

Russian charset problem

I've got Oracle database 8.1.7 and Oracle 9iAS installed on the same machine, Windows NT.
When I try to create form containing russian field names with Portal, it stores them (and then displays) in wrong charset. NLS_LANG parameters are all set to RUSSIAN_CIS.CL8MSWIN1251.
Does anyone know how to avoid this problem?

It's a bug.
When you create a new form edit it and add labels in Russian, before clicking Ok you should click on the root of the component tree (Form). In this case all Russian characters will be saved with proper encoding.

Large Charset Problem

Similar Messages

Maybe you are looking for